CN118329816B - Method and system for determining concentration of vanadium ions in positive electrolyte of all-vanadium redox flow battery - Google Patents
Method and system for determining concentration of vanadium ions in positive electrolyte of all-vanadium redox flow battery Download PDFInfo
- Publication number
- CN118329816B CN118329816B CN202410772461.4A CN202410772461A CN118329816B CN 118329816 B CN118329816 B CN 118329816B CN 202410772461 A CN202410772461 A CN 202410772461A CN 118329816 B CN118329816 B CN 118329816B
- Authority
- CN
- China
- Prior art keywords
- positive electrode
- electrode electrolyte
- vanadium
- concentration
- ion concentration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000003792 electrolyte Substances 0.000 title claims abstract description 131
- 229910001456 vanadium ion Inorganic materials 0.000 title claims abstract description 121
- 229910052720 vanadium Inorganic materials 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000003595 spectral effect Effects 0.000 claims abstract description 51
- 238000005259 measurement Methods 0.000 claims abstract description 38
- 238000002835 absorbance Methods 0.000 claims abstract description 27
- 238000000862 absorption spectrum Methods 0.000 claims abstract description 12
- 238000007599 discharging Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims description 45
- 238000002790 cross-validation Methods 0.000 claims description 14
- 230000003287 optical effect Effects 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 11
- 238000004448 titration Methods 0.000 claims description 7
- 238000001228 spectrum Methods 0.000 claims description 6
- 238000007405 data analysis Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000011156 evaluation Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000000870 ultraviolet spectroscopy Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 235000013405 beer Nutrition 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007865 diluting Methods 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005518 electrochemistry Effects 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 150000002500 ions Chemical class 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000007873 sieving Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- LEONUFNNVUYDNQ-UHFFFAOYSA-N vanadium atom Chemical compound [V] LEONUFNNVUYDNQ-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/30—Hydrogen technology
- Y02E60/50—Fuel cells
Landscapes
- Fuel Cell (AREA)
Abstract
The application provides a method and a system for measuring the concentration of vanadium ions in an anode electrolyte of an all-vanadium redox flow battery. The method comprises the following steps: extracting a spectral characteristic parameter x at the current moment from an absorbance spectrum of a positive electrode electrolyte during charging and discharging of the all-vanadium redox flow battery, wherein the spectral characteristic parameter x comprises absorbance x1 at one wavelength in a wavelength range of 400-450 nm, absorbance x2 at one wavelength in a wavelength range of 620-670nm and absorbance x3 at one wavelength in a wavelength range of 720-770 nm; and inputting the extracted spectral characteristic parameter x at the current moment into a trained positive electrode electrolyte vanadium ion concentration measurement model based on XGBoost algorithm to obtain a concentration parameter y at the current moment. The method and the system improve the detection accuracy and convenience of the concentration of vanadium ions and the SOC + of the positive electrode electrolyte.
Description
Technical Field
The application relates to the field of electrochemistry, but is not limited to, and in particular relates to a method and a system for determining the concentration of vanadium ions in positive electrolyte of an all-vanadium redox flow battery.
Background
The state of charge (SOC) of the positive electrode electrolyte of an all-vanadium flow battery is generally determined by the ratio of the tetravalent vanadium Ion V (IV) concentration and the pentavalent vanadium ion V (V) concentration in the positive electrode electrolyte. The concentration of vanadium ions can be obtained by titration, voltage method, optical method and other tests. The optical method can complete full-spectrum optical signal monitoring in a short time (the sampling rate can be lower than 5 ms), can collect mass data in unit time, is easy to realize online measurement and is gradually widely applied to signal analysis based on big data. However, for a system of a vanadium battery positive electrode electrolyte containing multiple light absorbing substances and having interaction of light absorbing particles (for example, tetravalent vanadium Ions V (IV) and pentavalent vanadium ions V (V) in the positive electrode electrolyte can interact to form valence states V 2O3 3+, etc.), the current optical measurement method based on lambert law or other single curve equation of concentration and absorbance cannot accurately and rapidly obtain the concentration of a single substance in the system.
Accordingly, there is a need for an improved method and system for accurately and conveniently determining the concentration of vanadium ions in the positive electrolyte of an all-vanadium redox flow battery in real time, and calculating the state of charge therefrom.
Disclosure of Invention
The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.
In one aspect, the application provides a method for determining the concentration of vanadium ions in the positive electrode electrolyte of an all-vanadium redox flow battery, which is characterized by comprising the following steps:
S1: extracting a spectral characteristic parameter x at the current moment from an absorbance spectrum of the positive electrode electrolyte during charging and discharging of the all-vanadium redox flow battery, wherein the spectral characteristic parameter x comprises absorbance x1 at one wavelength in a wavelength range of 400-450 nm, absorbance x2 at one wavelength in a wavelength range of 620-670nm and absorbance x3 at one wavelength in a wavelength range of 720-770 nm;
S2: inputting the extracted spectral characteristic parameter x at the current moment into a trained positive electrode electrolyte vanadium ion concentration determination model based on XGBoost algorithm to obtain a concentration parameter y at the current moment, wherein the concentration parameter y comprises the concentration C v(IV) y1 of tetravalent vanadium ions and the concentration C v(V) y2 of pentavalent vanadium ions in the positive electrode electrolyte;
the positive electrode electrolyte vanadium ion concentration determination model is trained in the following manner:
Extracting spectral characteristic parameters x1-x3 of a plurality of historical moments and obtaining concentration parameters y1-y2 of the historical moments;
inputting the spectral characteristic parameters x1-x3 as input data into the positive electrode electrolyte vanadium ion concentration measurement model, and training the positive electrode electrolyte vanadium ion concentration measurement model by taking the concentration parameters y1-y2 as expected output data of the positive electrode electrolyte vanadium ion concentration measurement model;
And when the actual output data of the positive electrode electrolyte vanadium ion concentration measuring model meets the set training termination condition, stopping training the positive electrode electrolyte vanadium ion concentration measuring model to obtain the trained positive electrode electrolyte vanadium ion concentration measuring model.
In one embodiment, extracting the spectral characteristic parameter x at the current moment comprises accessing a spectrophotometer in an anode electrolyte circulation system of the all-vanadium redox flow battery and performing real-time full-spectrum scanning in a charging and discharging process; and extracting the spectral feature parameters x1-x3 using a signal data analysis tool.
In an embodiment, obtaining the concentration parameter at the current time includes collecting a sample of the positive electrode electrolyte at the current time, and measuring the concentration C v(IV) y1 of tetravalent vanadium ions and the concentration C v(V) y2 of pentavalent vanadium ions in the positive electrode electrolyte by using an electrochemical titration method.
In one embodiment, training the positive electrode electrolyte vanadium ion concentration determination model includes constructing 80% of data in the extracted spectral feature parameters x1-x3 and the obtained concentration parameters y1-y2 into a training set, and 20% of data into a test set.
In one embodiment, training the positive electrode electrolyte vanadium ion concentration determination model comprises performing K-fold cross validation on one specified parameter in an initial model by adopting data of the training set, and screening out an optimal value of the specified parameter; substituting the screened optimal value of the specified parameter, continuing to carry out K-fold cross validation on the other specified parameter, and screening the optimal value of the specified parameter; and (5) until all the specified parameters complete K-fold cross validation, obtaining the current optimal model.
In one embodiment, the K-fold cross-validation is five-fold cross-validation.
In an embodiment, the specified parameters include n_ estimators, max _depth, min_child_ weight, gamma, subsample, colsample _ bytree, and learning_rate.
In an embodiment, training the positive electrode electrolyte vanadium ion concentration measurement model further includes continuing to train the optimal model using the data of the training set and evaluating the optimal model using the data of the test set until a set training termination condition is met.
In another aspect, a system for determining the concentration of vanadium ions in a positive electrode electrolyte of an all-vanadium redox flow battery is provided, comprising:
The optical measurement device is connected with the positive electrolyte of the all-vanadium redox flow battery and is used for carrying out real-time full spectrum scanning on the positive electrolyte during the charge and discharge of the all-vanadium redox flow battery to obtain an absorbance spectrum at the current moment;
The all-vanadium redox flow battery management system is connected with the optical measurement device and is provided with a trained positive electrode electrolyte vanadium ion concentration measurement model based on XGBoost algorithm, and is configured to input a spectral characteristic parameter x at the current moment extracted from an absorbance spectrum at the current moment into the trained positive electrode electrolyte vanadium ion concentration measurement model to obtain a concentration parameter y at the current moment, wherein the spectral characteristic parameter x is absorbance x1 at one wavelength in a wavelength range of 400-450 nm, absorbance x2 at one wavelength in a wavelength range of 620-670nm and absorbance x3 at one wavelength in a wavelength range of 720-770 nm; the concentration parameter y comprises the concentration C v(IV) y1 of tetravalent vanadium ions and the concentration C v(V) y2 of pentavalent vanadium ions in the positive electrode electrolyte;
wherein the positive electrode electrolyte vanadium ion concentration determination model is trained in the following manner:
Extracting spectral characteristic parameters x1-x3 of a plurality of historical moments and obtaining concentration parameters y1-y2 of the historical moments;
inputting the spectral characteristic parameters x1-x3 as input data into the positive electrode electrolyte vanadium ion concentration measurement model, and training the positive electrode electrolyte vanadium ion concentration measurement model by taking the concentration parameters y1-y2 as expected output data of the positive electrode electrolyte vanadium ion concentration measurement model;
And when the actual output data of the positive electrode electrolyte vanadium ion concentration measuring model meets the set training termination condition, stopping training the positive electrode electrolyte vanadium ion concentration measuring model to obtain the trained positive electrode electrolyte vanadium ion concentration measuring model.
In yet another aspect, a method for determining the state of charge SOC + of a positive electrode electrolyte of an all-vanadium redox flow battery is provided, comprising the steps of:
S1: extracting a spectral characteristic parameter x at the current moment from an absorbance spectrum of the positive electrode electrolyte during charging and discharging of the all-vanadium redox flow battery, wherein the spectral characteristic parameter x comprises absorbance x1 at one wavelength in a wavelength range of 400-450 nm, absorbance x2 at one wavelength in a wavelength range of 620-670nm and absorbance x3 at one wavelength in a wavelength range of 720-770 nm;
S2: inputting the extracted spectral characteristic parameter x at the current moment into a trained positive electrode electrolyte vanadium ion concentration determination model based on XGBoost algorithm to obtain a concentration parameter y at the current moment, wherein the concentration parameter y comprises the concentration of tetravalent vanadium ions in the positive electrode electrolyte
Concentration of C v(IV) y1 and pentavalent vanadium ion C v(V) y2; and
S3: the positive electrode electrolyte SOC + at the present time is calculated according to the following formula I:
（I）；
the positive electrode electrolyte vanadium ion concentration determination model is trained in the following manner:
Extracting spectral characteristic parameters x1-x3 of a plurality of historical moments and obtaining concentration parameters y1-y2 of the historical moments;
inputting the spectral characteristic parameters x1-x3 as input data into the positive electrode electrolyte vanadium ion concentration measurement model, and training the positive electrode electrolyte vanadium ion concentration measurement model by taking the concentration parameters y1-y2 as expected output data of the positive electrode electrolyte vanadium ion concentration measurement model;
And when the actual output data of the positive electrode electrolyte vanadium ion concentration measuring model meets the set training termination condition, stopping training the positive electrode electrolyte vanadium ion concentration measuring model to obtain the trained positive electrode electrolyte vanadium ion concentration measuring model.
The inventor of the application finds that the spectrograms in three specific wavelength ranges, namely the wavelength range of 400-450 nm, the wavelength range of 620-670nm and the wavelength range of 720-770nm, can reflect the change of the concentration of ions existing in the whole positive electrode electrolyte during the charge and discharge of the all-vanadium redox flow battery, and constructs and trains a positive electrode electrolyte vanadium ion concentration measurement model based on the change rule. The measurement accuracy of the trained positive electrode measurement model is obviously higher than that of a curve regression equation based on the lambert beer law or other existing concentration and absorbance.
According to the application, a machine learning model based on XGBoost algorithm is combined with the optical fiber spectrophotometer, so that the real-time in-situ accurate characterization of the vanadium ion concentration and the state of charge (SOC) of the positive electrode electrolyte of the all-vanadium redox flow battery is realized, and the problem of inaccurate measurement caused by the influence of factors such as covalent state substances and the like in the positive electrode electrolyte in the prior art is avoided.
Compared with the prior art that the step of diluting the electrolyte sample and then testing is needed for extracting the optical characteristics of the electrolyte, the method does not need to carry out the dilution step, is convenient and time-saving, and is beneficial to the automatic implementation of the detection method and system. In addition, the method can simultaneously obtain the respective accurate concentration values of two vanadium ions in the positive electrode electrolyte based on specific input parameters.
The method and the system can detect the concentration of vanadium ions in the positive electrode electrolyte in a wide range, for example, the concentration is close to 0 or the concentration is as high as 2.5mol/l, the detection sensitivity is high, and the output result is rapid.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application. Other advantages of the application may be realized and attained by the structure particularly pointed out in the written description.
Drawings
The accompanying drawings are included to provide an understanding of the principles of the application, and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain, without limitation, the principles of the application.
FIG. 1 is a flow chart of a method for determining the concentration of vanadium ions in the positive electrode electrolyte of an all-vanadium redox flow battery according to an exemplary embodiment of the present application;
FIG. 2 is a graph showing the variation of the evaluation index with the parameters provided in example 1 of the present application; and
Fig. 3 shows the predicted results of the initial model and the final model provided in example 1 of the present application compared with the actual values.
Detailed Description
The following describes embodiments of the present application in detail for the purpose of making the objects, technical solutions and advantages of the present application more apparent. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be arbitrarily combined with each other.
The application provides a method for determining the concentration of vanadium ions in an anode electrolyte of an all-vanadium redox flow battery, which comprises the following steps:
S1: extracting a spectral characteristic parameter x at the current moment from an absorbance spectrum of the positive electrode electrolyte during charging and discharging of the all-vanadium redox flow battery, wherein the spectral characteristic parameter x comprises absorbance x1 at one wavelength in a wavelength range of 400-450 nm, absorbance x2 at one wavelength in a wavelength range of 620-670nm and absorbance x3 at one wavelength in a wavelength range of 720-770 nm;
S2: inputting the extracted spectral characteristic parameter x at the current moment into a trained positive electrode electrolyte vanadium ion concentration determination model based on XGBoost algorithm to obtain a concentration parameter y at the current moment, wherein the concentration parameter y comprises the concentration C v(IV) y1 of tetravalent vanadium ions and the concentration C v(V) y2 of pentavalent vanadium ions in the positive electrode electrolyte;
the positive electrode electrolyte vanadium ion concentration determination model is trained in the following manner:
Extracting spectral characteristic parameters x1-x3 of a plurality of historical moments and obtaining concentration parameters y1-y2 of the historical moments;
inputting the spectral characteristic parameters x1-x3 as input data into the positive electrode electrolyte vanadium ion concentration measurement model, and training the positive electrode electrolyte vanadium ion concentration measurement model by taking the concentration parameters y1-y2 as expected output data of the positive electrode electrolyte vanadium ion concentration measurement model;
And when the actual output data of the positive electrode electrolyte vanadium ion concentration measuring model meets the set training termination condition, stopping training the positive electrode electrolyte vanadium ion concentration measuring model to obtain the trained positive electrode electrolyte vanadium ion concentration measuring model.
In the present application, the "history time" may represent a certain time before training, and the spectral feature parameter x at that time is extracted and the concentration parameter y at that time is obtained.
Fig. 1 is a flowchart of a method for determining the concentration of vanadium ions in a positive electrode electrolyte of an all-vanadium redox flow battery according to an exemplary embodiment of the present application. Fig. 1 shows that the method may comprise step S1 and step S2. Step S1 may include extracting a spectral feature parameter x at the current time from an absorbance spectrum of the positive electrode electrolyte during charging and discharging of the all-vanadium redox flow battery. The application selects specific spectral characteristic parameters x1-x3. Step S2 may include inputting the extracted spectral feature parameter x at the current time into a trained positive electrode electrolyte vanadium ion concentration determination model based on XGBoost algorithm, to obtain a concentration parameter at the current time. The positive electrode electrolyte vanadium ion concentration determination model can be trained in advance, and each time the positive electrode electrolyte vanadium ion concentration of the all-vanadium redox flow battery needs to be determined, the steps S1 and S2 are executed based on the trained positive electrode electrolyte vanadium ion concentration determination model.
In the application, the spectrophotometer can be an ultraviolet-visible spectrophotometer, and the ultraviolet-visible spectrophotometer can carry out real-time full-spectrum scanning on the charging and discharging process of the all-vanadium redox flow battery. The uv-vis spectrophotometer in the present application may be a commercially available instrument such as avants, the wavelength range of which is: 200-1100 nm, sampling frequency: 20 Hz.
In the present application, the signal data analysis tool may employ a Python signal data analysis tool SciPy.
In the application, the concentration of vanadium ions can be obtained by an electrochemical titration method, and the electrochemical titration method can be carried out by adopting a standard NB/T42006-2013. The electrochemical titration method has the most accurate test result, so the obtained concentration data is particularly beneficial to the accuracy of the subsequent model construction.
The initial model is evaluated and verified by using a test set in the application, and the following three indexes can be used for representing the difference between actual output data and expected output data of the positive electrode electrolyte vanadium ion concentration measurement model, wherein the three indexes are as follows: the determinable coefficient R 2, the mean square error MSE and the average absolute error MAE:
In the above-mentioned formula(s), Representing model predictive value, y i For actual measured concentration parameter values , Represents the average value of the y values of the concentration parameter actually measured, and N represents the number of predicted samples.
In the present application, the condition for model training termination may include R 2 not lower than 0.995 or MSE not higher than 0.0013. In addition, the change in mean absolute error MAE can see if the training is towards model optimization.
The model training method for detecting the concentration of vanadium ions and the state of charge (SOC +) of the positive electrode electrolyte of the all-vanadium redox flow battery in real time is specifically described below with reference to an embodiment.
Example 1
Real-time absorbance spectra were monitored and collected during the charge and discharge of the all-vanadium redox flow battery system using an ultraviolet-visible spectrophotometer system (avants, wavelength range: 350-1050 nm, sampling frequency: 20 Hz) on the positive electrolyte line. The spectral data is analyzed using Python signal data analysis tool SciPy, from which the spectral feature data at the current time is extracted: the absorbance at wavelengths 415 nm, 660 nm and 760 nm is used as the extracted spectral feature values x1, x2 and x3, namely, the input data of the positive electrode electrolyte vanadium ion concentration measurement model. And (3) sampling the electrolyte at the current moment, and measuring the concentration of tetravalent vanadium Ions V (IV) and pentavalent vanadium ions V (V) at the current moment according to the electrochemical titration method described by NB/T42006-2013, wherein the concentrations are used as the output y1 and y2 of the positive electrode electrolyte vanadium ion concentration measuring model. 1000 samples were collected containing spectral feature parameter x and concentration parameter y. Data samples were scaled 4:1 is divided into training and testing sets.
And establishing an initial determination model of the vanadium ion concentration of the positive electrolyte based on XGBoost algorithm, and performing binary regression. The initial model parameters are ：n_estimators：100,max_depth：5,min_child_weight：5,gamma：0,subsample：0.8,colsample_bytree：0.8,learning_rate：0.1, the remaining parameters using default values. After training a model by using the training set, bringing the spectral characteristic parameter x of the test set into the model, comparing the predicted result with the concentration parameter y of the test set, and calculating an evaluation index determinable coefficient R 2, a mean square error MSE and an average absolute error MAE.
The training process of the model may include determining n_ estimators, max_depth, min_child_weight, gamma, subsamples, colsample _ bytree, and learning_rate as specified parameters; the data of the training set are sequentially subjected to optimal parameter screening by using five-fold cross validation, wherein the optimal value of one appointed parameter in the initial model is screened out by performing K-fold cross validation on the appointed parameter; substituting the screened optimal value of the specified parameter, continuing to carry out K-fold cross validation on the other specified parameter, and screening the optimal value of the specified parameter; until all the appointed parameters finish K-fold cross validation; specifically, the screening range of n_ estimators is 50-1000, and the screened optimal parameter is 560; the screening range of max_depth is 1-10, and the screened optimal parameter is 6; the screening range of the min_child_weight is 1-10, and the screened optimal parameter is 6; the optimal parameter after gamma screening is still 0; optimal parameters after sieving the subsamples and colsample _ bytree are 1; the optimal parameter after the learning_rate screening was 0.46. And after the optimal value of the specified parameter is screened out, updating the parameter value of the initial model to obtain the current optimal model. The result of model evaluation of the determinable coefficient R 2, the mean square error MSE and the mean absolute error MAE after the initial model and the update of the parameters is shown in fig. 2. As can be seen from fig. 2, the three evaluation indicators are changed substantially in a better direction after each update of the parameters.
This optimal model is then trained and evaluated using the same training set and test set data. Fig. 3 shows the predicted result in comparison with the actual value. As can be seen from fig. 3, the dotted line is the state when the prediction is 100% accurate, the upper graph in fig. 3 is the initial model prediction result data, and the lower graph in fig. 3 is the final model (current optimal model) prediction result data. As can be seen from fig. 3, the prediction of the final model enables more data points to fall on the dotted line, indicating that the prediction effect is improved.
The trained model is imported into a battery management system BMS.
The battery management system BMS inputs the optical parameter characteristic values into the trained model by extracting the optical parameter characteristic values from the optical signals measured by the ultraviolet spectrophotometer, so that the concentrations of tetravalent vanadium Ions V (IV) and pentavalent vanadium ions V (V) are obtained and output, namely: and outputting and obtaining the concentration of tetravalent vanadium Ions V (IV) and pentavalent vanadium ions V (V) in the positive electrode electrolyte from the battery management system BMS. Further, the state of charge SOC + of the positive electrode electrolyte may be further calculated based on the concentration value.
Although the embodiments of the present application are described above, the embodiments are only used for facilitating understanding of the present application, and are not intended to limit the present application. Any person skilled in the art can make any modification and variation in form and detail without departing from the spirit and scope of the present disclosure, but the scope of the present disclosure is still subject to the scope of the present disclosure as defined by the appended claims.
Claims (10)
1. A method for determining the concentration of vanadium ions in a positive electrode electrolyte of an all-vanadium redox flow battery, comprising the following steps:
S1: extracting a spectral characteristic parameter x at the current moment from an absorbance spectrum of the positive electrode electrolyte during charging and discharging of the all-vanadium redox flow battery, wherein the spectral characteristic parameter x comprises absorbance x1 at one wavelength in a wavelength range of 400-450 nm, absorbance x2 at one wavelength in a wavelength range of 620-670nm and absorbance x3 at one wavelength in a wavelength range of 720-770 nm;
S2: inputting the extracted spectral characteristic parameter x at the current moment into a trained positive electrode electrolyte vanadium ion concentration determination model based on XGBoost algorithm to obtain a concentration parameter y at the current moment, wherein the concentration parameter y comprises the concentration C v(IV) y1 of tetravalent vanadium ions and the concentration C v(V) y2 of pentavalent vanadium ions in the positive electrode electrolyte;
the positive electrode electrolyte vanadium ion concentration determination model is trained in the following manner:
Extracting spectral characteristic parameters x1-x3 of a plurality of historical moments and obtaining concentration parameters y1-y2 of the historical moments;
inputting the spectral characteristic parameters x1-x3 as input data into the positive electrode electrolyte vanadium ion concentration measurement model, and training the positive electrode electrolyte vanadium ion concentration measurement model by taking the concentration parameters y1-y2 as expected output data of the positive electrode electrolyte vanadium ion concentration measurement model;
And when the actual output data of the positive electrode electrolyte vanadium ion concentration measuring model meets the set training termination condition, stopping training the positive electrode electrolyte vanadium ion concentration measuring model to obtain the trained positive electrode electrolyte vanadium ion concentration measuring model.
2. The method of claim 1, wherein extracting the spectral feature parameter x at the current moment comprises accessing a spectrophotometer in a positive electrolyte circulation system of the all-vanadium redox flow battery and performing real-time full-spectrum scanning in a charging and discharging process; and extracting the spectral feature parameters x1-x3 using a signal data analysis tool.
3. The method of claim 1, wherein obtaining the concentration parameter at the current time comprises taking a sample of the positive electrode electrolyte at the current time and measuring the concentration of tetravalent vanadium ions C v(IV) y1 and the concentration of pentavalent vanadium ions C v(V) y2 in the positive electrode electrolyte using an electrochemical titration method.
4. The method of claim 1, wherein training the positive electrolyte vanadium ion concentration determination model comprises constructing 80% of data in the extracted spectral feature parameters x1-x3 and the obtained concentration parameters y1-y2 as a training set and 20% of data as a test set.
5. The method of claim 4, wherein training the positive electrolyte vanadium ion concentration determination model comprises K-fold cross-validation of one specified parameter in an initial model using data of the training set, and screening out an optimal value of the specified parameter; substituting the screened optimal value of the specified parameter, continuing to carry out K-fold cross validation on the other specified parameter, and screening the optimal value of the specified parameter; and (5) until all the specified parameters complete K-fold cross validation, obtaining the current optimal model.
6. The method of claim 5, wherein the K-fold cross-validation is a five-fold cross-validation.
7. The method of claim 5, wherein the specified parameters include n_ estimators, max _depth, min_child_ weight, gamma, subsample, colsample _ bytree, and learning_rate.
8. The method of claim 5, wherein training the positive electrolyte vanadium ion concentration determination model further comprises continuing to train the optimal model using the data of the training set and evaluating the optimal model using the data of the test set until a set training termination condition is met.
9. A system for determining the concentration of vanadium ions in a positive electrolyte of an all-vanadium redox flow battery, comprising:
The optical measurement device is connected with the positive electrolyte of the all-vanadium redox flow battery and is used for carrying out real-time full spectrum scanning on the positive electrolyte during the charge and discharge of the all-vanadium redox flow battery to obtain an absorbance spectrum at the current moment;
The all-vanadium redox flow battery management system is connected with the optical measurement device and is provided with a trained positive electrode electrolyte vanadium ion concentration measurement model based on XGBoost algorithm, and is configured to input a spectral characteristic parameter x at the current moment extracted from an absorbance spectrum at the current moment into the trained positive electrode electrolyte vanadium ion concentration measurement model to obtain a concentration parameter y at the current moment, wherein the spectral characteristic parameter x comprises absorbance x1 at one wavelength in a 400-450 nm wavelength range, absorbance x2 at one wavelength in a 620-670nm wavelength range and absorbance x3 at one wavelength in a 720-770nm wavelength range; the concentration parameter y comprises the concentration C v(IV) y1 of tetravalent vanadium ions and the concentration C v(V) y2 of pentavalent vanadium ions in the positive electrode electrolyte;
wherein the positive electrode electrolyte vanadium ion concentration determination model is trained in the following manner:
Extracting spectral characteristic parameters x1-x3 of a plurality of historical moments and obtaining concentration parameters y1-y2 of the historical moments;
inputting the spectral characteristic parameters x1-x3 as input data into the positive electrode electrolyte vanadium ion concentration measurement model, and training the positive electrode electrolyte vanadium ion concentration measurement model by taking the concentration parameters y1-y2 as expected output data of the positive electrode electrolyte vanadium ion concentration measurement model;
And when the actual output data of the positive electrode electrolyte vanadium ion concentration measuring model meets the set training termination condition, stopping training the positive electrode electrolyte vanadium ion concentration measuring model to obtain the trained positive electrode electrolyte vanadium ion concentration measuring model.
10. A method for determining the state of charge SOC + of a positive electrode electrolyte of an all-vanadium redox flow battery, comprising the steps of:
S1: extracting a spectral characteristic parameter x at the current moment from an absorbance spectrum of the positive electrode electrolyte during charging and discharging of the all-vanadium redox flow battery, wherein the spectral characteristic parameter x comprises absorbance x1 at one wavelength in a wavelength range of 400-450 nm, absorbance x2 at one wavelength in a wavelength range of 620-670nm and absorbance x3 at one wavelength in a wavelength range of 720-770 nm;
s2: inputting the extracted spectral characteristic parameter x at the current moment into a trained positive electrode electrolyte vanadium ion concentration determination model based on XGBoost algorithm to obtain a concentration parameter y at the current moment, wherein the concentration parameter y comprises the concentration C v(IV) y1 of tetravalent vanadium ions and the concentration C v(V) y2 of pentavalent vanadium ions in the positive electrode electrolyte; and
S3: the positive electrode electrolyte SOC + at the present time is calculated according to the following formula I:
（I）；
the positive electrode electrolyte vanadium ion concentration determination model is trained in the following manner:
Extracting spectral characteristic parameters x1-x3 of a plurality of historical moments and obtaining concentration parameters y1-y2 of the historical moments;
inputting the spectral characteristic parameters x1-x3 as input data into the positive electrode electrolyte vanadium ion concentration measurement model, and training the positive electrode electrolyte vanadium ion concentration measurement model by taking the concentration parameters y1-y2 as expected output data of the positive electrode electrolyte vanadium ion concentration measurement model;
And when the actual output data of the positive electrode electrolyte vanadium ion concentration measuring model meets the set training termination condition, stopping training the positive electrode electrolyte vanadium ion concentration measuring model to obtain the trained positive electrode electrolyte vanadium ion concentration measuring model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410772461.4A CN118329816B (en) | 2024-06-14 | 2024-06-14 | Method and system for determining concentration of vanadium ions in positive electrolyte of all-vanadium redox flow battery |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410772461.4A CN118329816B (en) | 2024-06-14 | 2024-06-14 | Method and system for determining concentration of vanadium ions in positive electrolyte of all-vanadium redox flow battery |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118329816A CN118329816A (en) | 2024-07-12 |
| CN118329816B true CN118329816B (en) | 2024-08-20 |
Family
ID=91780681
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410772461.4A Active CN118329816B (en) | 2024-06-14 | 2024-06-14 | Method and system for determining concentration of vanadium ions in positive electrolyte of all-vanadium redox flow battery |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118329816B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101995385A (en) * | 2009-08-27 | 2011-03-30 | 中国科学院金属研究所 | Ultraviolet quantitative determination method for concentration of vanadium battery positive electrolyte and application thereof |
| CN107422267A (en) * | 2017-04-10 | 2017-12-01 | 上海电气集团股份有限公司 | The SOC detection means and method of all-vanadium flow battery |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102016117604A1 (en) * | 2016-09-19 | 2018-03-22 | Thyssenkrupp Ag | Determination of the charge state of an all-vanadium redox flow battery by means of UV / Vis measurement |
| CN110857911B (en) * | 2018-08-24 | 2021-11-26 | 江苏泛宇能源有限公司 | Method for testing electrolyte balance degree of portable all-vanadium redox flow battery |
| KR102587671B1 (en) * | 2021-07-05 | 2023-10-12 | 한국전력공사 | Vanadium redox flow battery charge/discharge monitoring device and real-time measuring method |
| CN115133081B (en) * | 2022-08-29 | 2022-12-30 | 液流储能科技有限公司 | Method for testing positive electrode charging state and vanadium ion total concentration in all-vanadium redox flow battery |
| CN117192392A (en) * | 2023-09-06 | 2023-12-08 | 江苏泛锐思智能技术有限公司 | State of charge monitoring method and device for all-vanadium liquid flow energy storage system |
| CN118011245A (en) * | 2024-02-02 | 2024-05-10 | 上海碳动新能源科技有限公司 | Method, device and system for detecting state of charge of battery, electronic equipment and medium |
-
2024
- 2024-06-14 CN CN202410772461.4A patent/CN118329816B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101995385A (en) * | 2009-08-27 | 2011-03-30 | 中国科学院金属研究所 | Ultraviolet quantitative determination method for concentration of vanadium battery positive electrolyte and application thereof |
| CN107422267A (en) * | 2017-04-10 | 2017-12-01 | 上海电气集团股份有限公司 | The SOC detection means and method of all-vanadium flow battery |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118329816A (en) | 2024-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107703132B (en) | Reaction curve abnormity processing method and device, biochemical analyzer and storage medium | |
| CN101988894B (en) | Method for determining a parameter of liquid sample | |
| KR102106775B1 (en) | Diagnosis method of battery using Deep Learning | |
| US20220146417A1 (en) | Method for constructing water quality index prediction models and method for monitoring water quality indexes | |
| CN103134767B (en) | Method for liquor quality identification through infrared spectrum revision | |
| CN114325403B (en) | Lithium ion battery life detection method and system based on electrochemical impedance spectrum test | |
| Zhang et al. | An on-line spectroscopic monitoring system for the electrolytes in vanadium redox flow batteries | |
| CN115236045A (en) | Method and device for determining chemical oxygen demand in water by using fluorescence spectrometry and storage medium | |
| CN110567937A (en) | Competitive self-adaptive heavy-weighted key data extraction method for Raman spectrum analysis of insulating oil | |
| DE102017126612A1 (en) | Measuring device for analyzing a measuring medium | |
| AU2020203717A1 (en) | Detecting a transient error in a body fluid sample | |
| CN115993344A (en) | Quality monitoring and analyzing system and method for near infrared spectrum analyzer | |
| CN111879709A (en) | Method and device for detecting spectral reflectivity of lake water body | |
| CN116467950A (en) | Unmanned aerial vehicle flight data anomaly detection method based on uncertain characterization | |
| CN118329816B (en) | Method and system for determining concentration of vanadium ions in positive electrolyte of all-vanadium redox flow battery | |
| CN117591978A (en) | Environment pollution detection method, system, equipment and storage medium | |
| CN110084420B (en) | Method for detecting total sugar, total acid and alcoholic strength of yellow water in strong aromatic Chinese spirit fermentation | |
| CN105954206B (en) | The measurement method and system of purple maize leaf anthocyanin content | |
| CN118329817B (en) | Method and system for determining concentration of vanadium ions in negative electrolyte of all-vanadium redox flow battery | |
| CN110879222A (en) | Construction method of quantitative analysis curve of furfural in transformer oil and detection method of furfural content | |
| CN105866065B (en) | Methenamine content analysis method in a kind of methenamine-acetum | |
| CN112257017B (en) | Unitary linear point-by-point analysis method, system and device for standardized residual error detection method | |
| CN111898314A (en) | Lake water body parameter detection method and device, electronic equipment and storage medium | |
| Iurilli | EIS applied to Li-Ion Batteries: Modelling Cell Aging for Comprehensive SoH Estimation | |
| CN114486805B (en) | Method for determining process parameters of hydrogen peroxide production process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |